Too much one-on-one PDF Print E-mail
Written by Doug Thonus   
Thursday, 14 May 2009 16:01

Does Ben Gordon go one on one too much?   This is a common complaint put forth by Gordon detractors.   Since I have the means, the time, and the intense desire to find out the truth I decided to research this topic by the numbers.


First, let’s look at the Bulls perimeter players themselves, here’s a list of the percentage of times each guy goes 1 on 1 followed by the NBA percentile in their effectiveness of doing so.

Bulls players isolation usage and effectiveness:
John Salmons 25.65% / 51%
Ben Gordon 20.34% / 69%
Derrick Rose 14.24% / 47%
Luol Deng 9.09% / 6%
Kirk Hinrich 9.02% / 52%

What does this mean?   Well first, it means supporting John Salmons to take over for Ben Gordon because Gordon goes one on one too much is ridiculous.   Salmons goes one on one more and is a lot worse at it.   It also means those who feel Rose is the best scorer in an isolation appear to be wrong.  His 47th percentile in the league makes him an average scorer one on one.  

Luol Deng should never have isolation opportunities as his numbers are scary bad, and Kirk Hinrich looks like he does a decent job for the limited chances he has.    This also backs up my assertion that the Bulls were better off giving the ball to Gordon at the end of a game than Rose as Gordon was a substantially better scorer in an isolation situation.  

Now, looking at these numbers you could certainly argue that Gordon stops the ball more than Hinrich, so those who feel Hinrich would facilitate better ball movement than Gordon would appear to be correct.   Whether that ball movement facilitates more scoring than Gordon is certainly debatable given how successful Gordon was in the isolation role though.

Looking at the Bulls players is fine and dandy, we can learn something about how the Bulls operated as a team, however, I don’t want to stop there.   Let’s take a look at how Gordon matches up with the rest of the league.    The following is a list of every team’s best perimeter scorer except for teams which have no real perimeter threats and only have guys who just shoot kick outs. The list contains 37 different players.   Of that group, Pierce, Martin, and Redd didn’t have a percentile.  I believe in the case of Redd/Martin it was due to not enough games played with Pierce it appeared his possessions weren’t logged correctly.   The list is ordered in terms of how frequently the players used isolation in their game.

Usage and effectiveness of isolation around the league:
Kobe Bryant: 32.45% / 84%
Joe Johnson: 30.66% / 72%
Carmelo Anthony 30.28% / 47%
Brandon Roy: 29.33% / 87%
Caron Butler: 28.68% / 58%
Corey Maggette: 28.52% / 83%
Tracy McGrady: 28.12% / 35%
Stephen Jackson: 27.55% / 73%
John Salmons: 25.65% / 51%
Baron Davis: 25.07% / 38%
Allen Iverson: 24.52% / 41%
Jamal Crawford: 23.6% / 43%
LeBron James: 22.78% / 69%
Paul Pierce: 22.22% / NA
Deron Williams: 22.2% / 88%
Dwyane Wade: 20.59% / 91%
Vince Carter: 20.45% / 73%
Ben Gordon : 20.34% / 69%
Kevin Martin: 19.38% / NA
JR Smith: 19.16% / 74%
Michael Redd: 19.08% / NA
Kevin Durant: 18.49% / 42%
Andre Iguodala: 17.58% / 40%
Randy Foye: 16.49% / 54%
Josh Howard: 16.45% / 43%
Russell Westbrook: 16.16% / 37%
Eric Gordon: 15.69 / 46%
Gerald Wallace: 15.17% / 46%
Manu Ginobili: 15.13% / 88%
Nate Robinson: 14.72% / 68%
OJ Mayo: 14.48% / 62%
Chris Paul: 13.17% / 62%
Danny Granger: 12.04% / 79%
Steve Nash: 10.61% / 94%
Rip Hamilton: 10.51% / 34%
Jason Terry: 7.77% / 55%
Ray Allen: 4.56% / 86%

Results:
In terms of how often the players used isolation play Gordon ranked 18 out of 37 which is basically right in the middle of the list.   In terms of effectiveness he ranked 14 out of 34 which is in the upper middle portion of the list.   In terms of overall scoring rate he ranked 14 out of 34 on this list.  

So is Gordon selfish?   It would not appear so.    His scoring rate and effectiveness both were greater than his use of isolation relative to his peers even if all three of the NA players were to have greater efficiency than Gordon as isolation players.    

It would appear that Gordon uses an appropriate amount of isolation plays for a big time scorer and for the effectiveness in which he performs in those situations relative to his peers and that complaints about Gordon using too much one-on-one are simply out of line with what the rest of the good scorers in the league use. 

It's also worth noting the majority of two guards whom most people would agree are clearly better than Gordon (Roy, Wade, Kobe, Joe Johnson) all use more isolation than Gordon, with only Wade being close to Gordon's rate while the other three far exceed his isolation usage.   Of the players most frequently compared to Gordon Ray Allen has almost no isolation usage (gets most of his shots as a spot up shooter or coming off screens) while Redd has a very similar rate of isolation usage.

What to take from these numbers is up to the individual.   They show me that Gordon definitely uses isolation plays quite a bit.   There is no denying that fact, and the complaints about Gordon using too much isolation clearly stem from the fact that he does go to the iso quite a bit.   However, the numbers also bear out that his usage of isolation is well within the bounds of what other quality shooting guards use especially given his effectiveness as an isolation player.   What is scariest is the amount of people clamouring for John Salmons to start at shooting guard who claim Ben Gordon is a ball stopper given that Salmons use a significantly higher percentage of isolation plays at a significantly lower rate of effectiveness.

 

(stats provided by synergysportstech)

Last Updated ( Thursday, 14 May 2009 16:42 )
 
Discuss (83 posts)
Re:Too much one-on-one
May 15 2009 14:06:53
Given how unlikely it is that Gordon see's more money than he turned down the last two off-seasons is everyone still going to blame Pax if he goes? Seems to me he was willing to put the best offer no the table and BG over-valued himself. Nothing wrong with that, players got the right to get paid. I just don't want to see Pax end up the villan in all of this.
#14766
Re:Too much one-on-one
May 15 2009 14:34:25
dougthonus wrote:

I guess you're right... not too inconsistent to be a #1 scorer...
Playoffs...
Game 1 - 35% FG%
Game 2 - 58% FG%
Game 3 - 39% FG%
Game 4 - 46% FG% - Hurt hamstring during game
Game 5 - 29% FG%
Game 6 - 29% FG%
Game 7 - 30% FG%


FG% isn't a particularly accurate way to judge consistency. Lists points per possession used for all his games.

A possession is when you shoot, get to the line, or turn the ball over:

Game 1: 20 points, 21 possessions
Game 2: 42 points, 28 possessions
Game 3: 15 points, 17 possessions
Game 4: 22 points, 25 possessions
Game 5: 26 points, 28 possessions
Game 6: 12 points, 18 possessions
Game 7: 33 points, 32 possessions
Total: 170 points, 169 possessions

He had one game where he was ultra efficient, one game where he was terribly inefficient, and five games where he was close to 1:1. He played with an injury which notably hampered him for three of the seven games. His scoring consistency was pretty his low water marks were 12 points and 15 points.

Let's look at Paul Pierce:
Game 1: 24 points, 28 possessions
Game 2: 18 points, 24 possessions
Game 3: 24 points, 17 possessions
Game 4: 29 points, 34 possessions
Game 5: 26 points, 26 possessions
Game 6: 22 points, 22 possessions
Game 7: 20 points, 24 possessions
Total: 163 points, 175

He had 1 game where he was ultra efficient, 4 gamers where he was moderately inefficient, and two games where he was 1:1, His point production was very consistent even if his scoring efficiency wasn't.

Let's look at Ray Allen:
Game 1: 4 points, 14 possessions
Game 2: 30 points, 21 possessions
Game 3: 18 points, 15 possessions
Game 4: 28 points, 23 possessions
Game 5: 10 points, 11 possessions
Game 6: 51 points, 38 possessions
Game 7: 23 points, 18 possessions
Total: 165 points, 140 possessions
Allen was great, outside of his game 1, he was basically awesome. He was consistently efficient, having very good efficiency in 5 of 7 games, 1 super dud, and 1 game where he was average. Of course his points were all over the board, as he has 2 very low point games.

Let's look at Derrick Rose:
Game 1: 36 points, 30 possessions
Game 2: 10 points, 13 possessions
Game 3: 9 points, 22 possessions
Game 4: 23 points, 26 possessions
Game 5: 14 points, 26 possessions
Game 6: 28 points, 33 possessions
Game 7: 18 points, 21 possessions
Total: 138 points, 171 possessions
Basically, his efficiency was god awful for most of the series with only one game of good efficiency all series. His scoring consistency was also pretty bad with games of 9, 10, and 14 points while he scored 36 and 28 in other games.

Let's look at John Salmons
Game 1: 12 points, 17 possessions
Game 2: 17 points, 21 possessions
Game 3: 14 points, 17 possessions
Game 4: 20 points, 19 possessions
Game 5: 17 points, 19 possessions
Game 6: 35 points, 26 possessions
Game 7: 12 points, 17 possessions
Total: 127 points, 136 possessions
Salmons played through a sore groin which may have effected his play (though he did not look as visibly hobbled as Ben after the hamstring injury). His efficiency was basically poor for the whole series except for game 6, his total point consistency was basically poor too as he had duds in three games and lower games two more with only 17 points.


And... 169 points in 170 shots.... and that's with what you called an ultra efficient game. So you take that away and you get 127 in 142 shots? Even without counting the 3 hamstring games he had 2 bad shooting games of the first 4... and his ultra efficient game went for not since we lost it. So of the 3 games we won in the series.... Gordon shot bad in those games... So maybe the winning for this team isn't so contengent on him being a top scorer for this team?


Overall Gordon's stats usually average out well, it's those extream peaks and valleys that keep him from being a SOLID night in and night out top option for a very good team. This is why you can't build around Gordon and expect to be an elite team... and it's exactly the reason why him coming off the bench can be a very important piece for an elite team... you always have that chance he's going to go off if your team is struggling unless for some reason you think most good teams can have their top players play bad a third of the time and the team still be a good team.

I'm not argueing that statistically he doesn't come out good. What I'm argueing is a GAME BY GAME basis... his scoring/shooting is just too much of a variable to be a leader of a consistantly good team. You build a system offensivly but have one of the main guys consistantly(every third or fourth game) shoot horrible...
#14770
Re:Too much one-on-one
May 15 2009 16:36:39
First, I thought it was a really interesting article, so thanks for taking the time to research and write it, Doug. I think we get hyper-critical of Ben, and really all the Bulls players, just because of how much we see them and because we really care when a mistake loses a game for us. I've had League Pass the last two seasons, and semi-regularly watching other games has made me appreciate that our guys are really pretty good. One game in particular I was watching with a friend and at the end Chris Paul made a truly horrible game/clock management decision that cost NO the game, and when I commented on it my friend just dismissed it as "everyone makes mistakes now and then" - the same friend who has been known to absolutely rant on a Bulls' player and demand demotion to the D League for a not-so-bad mistake. And that Bulls' player's mistake will then be brought up over and over to build a case for how much they need replaced. "Remember that one game when Ben tripped over his own feet? He's not worth a dime!" Doug's article shows that Ben is behaving as many big time scorers do, and he's doing it more effectively than most - trade him for someone else to slide in there, and odds are good we'll be looking at the same thing. (Especially in this offense Vinny has going...)

Second, I think this thread shows perfectly how difficult it is to get most of us to even consider things differently - no matter what info is put out there, we just interpret it as supporting our own viewpoint or dismiss it.
#14778
Re:Too much one-on-one
May 15 2009 17:02:10
AirP wrote:
Overall Gordon's stats usually average out well, it's those extream peaks and valleys that keep him from being a SOLID night in and night out top option for a very good team. This is why you can't build around Gordon and expect to be an elite team... and it's exactly the reason why him coming off the bench can be a very important piece for an elite team... you always have that chance he's going to go off if your team is struggling unless for some reason you think most good teams can have their top players play bad a third of the time and the team still be a good team.

I'm not argueing that statistically he doesn't come out good. What I'm argueing is a GAME BY GAME basis... his scoring/shooting is just too much of a variable to be a leader of a consistantly good team. You build a system offensivly but have one of the main guys consistantly(every third or fourth game) shoot horrible...


You know, I did a comparison of Ben to other top shooting guards on this very question not too long ago. I don't have access to all the fun stats Doug does - I just used points per game for the entire season to calculate the standard deviations for the top 7 scoring SGs as a measure of how much variation there is in the scoring of each from night to night. Ben came out pretty much right in line with the others. Taking standard deviation as a percentage of average, D-Wade and Kobe rates as most consistent, but Ben was just a bit behind Brandon Roy, Joe Johnson, and Andre Iguodala, and was substantially ahead of Vince Carter.

Ben was also right in range with the others (except D-Wade and Kobe, again, who never scored less than 10) when I looked at number of single-digit-points nights throughout the season, and minimum points scored in at least 80% of games compared to average points (what you can "pencil him in for" on any given night).

So my conclusion was that Ben is equally consistent/inconsistent (whichever way you want to view it) as the other top-scoring SGs in the league. When you consider the variables in his season - new rookie coach, rookie PG, big trade during the season, other key guys going out with injuries - I think it's reasonable to say Ben is equal to his peers.

www.blogabull.com/2009/4/25/853143/concerning-the-alleged
#14780
Re:Too much one-on-one
May 15 2009 18:28:39
wjb1492 wrote:
AirP wrote:
Overall Gordon's stats usually average out well, it's those extream peaks and valleys that keep him from being a SOLID night in and night out top option for a very good team. This is why you can't build around Gordon and expect to be an elite team... and it's exactly the reason why him coming off the bench can be a very important piece for an elite team... you always have that chance he's going to go off if your team is struggling unless for some reason you think most good teams can have their top players play bad a third of the time and the team still be a good team.

I'm not argueing that statistically he doesn't come out good. What I'm argueing is a GAME BY GAME basis... his scoring/shooting is just too much of a variable to be a leader of a consistantly good team. You build a system offensivly but have one of the main guys consistantly(every third or fourth game) shoot horrible...


You know, I did a comparison of Ben to other top shooting guards on this very question not too long ago. I don't have access to all the fun stats Doug does - I just used points per game for the entire season to calculate the standard deviations for the top 7 scoring SGs as a measure of how much variation there is in the scoring of each from night to night. Ben came out pretty much right in line with the others. Taking standard deviation as a percentage of average, D-Wade and Kobe rates as most consistent, but Ben was just a bit behind Brandon Roy, Joe Johnson, and Andre Iguodala, and was substantially ahead of Vince Carter.

Ben was also right in range with the others (except D-Wade and Kobe, again, who never scored less than 10) when I looked at number of single-digit-points nights throughout the season, and minimum points scored in at least 80% of games compared to average points (what you can "pencil him in for" on any given night).

So my conclusion was that Ben is equally consistent/inconsistent (whichever way you want to view it) as the other top-scoring SGs in the league. When you consider the variables in his season - new rookie coach, rookie PG, big trade during the season, other key guys going out with injuries - I think it's reasonable to say Ben is equal to his peers.

www.blogabull.com/2009/4/25/853143/concerning-the-alleged


If you take more into consideration then just "ponits"... lets say like... GameScore which takes in account a lot of different "other" areas, the reason why the others seem to be looked upon better becomes apparent...

Game Score; the formula is PTS + 0.4 * FG - 0.7 * FGA - 0.4*(FTA - FT) + 0.7 * ORB + 0.3 * DRB + STL + 0.7 * AST + 0.7 * BLK - 0.4 * PF - TOV. Game Score was created by John Hollinger to give a rough measure of a player's productivity for a single game. The scale is similar to that of points scored, i.e., 40 is an outstanding performance, 10 is an average performance, etc.

Games below 10.0 for their gamescore this year...
Ben Gordon 27 games - Nearly a third of his games
Joe Johnson 20 games
Dwayne Wade 4 games
Kobe 8 games
Roy 10 games
Iggy 17 games
V.Carter 22 games

I guess if all you care about is "got mine" type of players, Gordon does "get his", he just isn't near the player the other guys are... even with V.Carter on the backside of his career.
#14790
Re:Too much one-on-one
May 16 2009 00:53:47
AirP wrote:

If you take more into consideration then just "ponits"... lets say like... GameScore which takes in account a lot of different "other" areas, the reason why the others seem to be looked upon better becomes apparent...

Game Score; the formula is PTS + 0.4 * FG - 0.7 * FGA - 0.4*(FTA - FT) + 0.7 * ORB + 0.3 * DRB + STL + 0.7 * AST + 0.7 * BLK - 0.4 * PF - TOV. Game Score was created by John Hollinger to give a rough measure of a player's productivity for a single game. The scale is similar to that of points scored, i.e., 40 is an outstanding performance, 10 is an average performance, etc.

Games below 10.0 for their gamescore this year...
Ben Gordon 27 games - Nearly a third of his games
Joe Johnson 20 games
Dwayne Wade 4 games
Kobe 8 games
Roy 10 games
Iggy 17 games
V.Carter 22 games

I guess if all you care about is "got mine" type of players, Gordon does "get his", he just isn't near the player the other guys are... even with V.Carter on the backside of his career.


See, but now you're changing the question. I looked at points only and compared him to that group because a question had come up in the context of scoring - Ben is an elite scorer, so that's best group to compare him to when discussing variations in points scored. Obviously, when you start adding other stats into the equation, Ben is going to rate differently than those other players. I would absolutely agree that he's a bit farther down the list as an all-around player - that is, he scores roughly "as much" as other 20+ ppg guy, but many of them contribute more in other categories. And comparing Ben to max contract type guys for all-around game is no longer the correct comparison group - if Ben had the all-around talent of these other guys we wouldn't be arguing about whether he's worth $8M or $10M per year.

If you want to get to the issue of variation, you can't just look at how many games fall below a certain constant threshhold. Ben's average gamescore and st.dev. for the season were 13.59 and 7.68 (I used gamescores in a teamwide analysis of player "great games" and "awful games" and how those related to team record in another post over on BaB, so have those for Bulls players but not anyone else in the NBA), so it is absolutely to be expected that he'd have more games below 10 than another player who's average gamescore was higher to begin with - but at this point you haven't addressed consistency at all - is that 7.68 standard deviation excessively high, or is it right in line with other players of Ben's quality?
#14813
Too much one-on-one
May 16 2009 13:05:21
This is a classic example of the dangers of over-focusing on a narrow set of statistics. In every sport, countless people seek the magic number – or numbers – to be used as a shortcut. But the truth is that no narrow set of numbers can ever substitute for the more sophisticated analyses necessary to make good personnel decisions. And, in many cases, such as this one, the numbers are actually misleading.

With regards to this particular statistical approach, it is obviously flawed, and is a ridiculous way to compare Gordon and Salmons (or any other players). According to the above list, Salmons, Randy Foy and Nate Robinson are all more "effective" one-on-one players than Carmelo Anthony. Do I really have to elaborate further?

Furthermore, I'd sure like to know how "one-on-one" is defined. Is there necessarily dribbling involved, or is a contested jump shot considered as well? In other words, let's see some definitions so that we can take into account how the posted (and clearly flawed) numbers were generated.

The way that players perform is also obviously dependent in large part on how they are coached. In the context of the Bulls, VDN was so atrocious that to judge players on their raw numbers (even accurate and meaningful ones) while they played in such a poorly designed offensive is silly. In other words, the biggest single step the Bulls could take would be to replace VDN, irrespective of the players in uniform. A better coach, for example, would not resort to predictably having Gordon go one-on-one (or two) at the end of tight games. The point being that Gordon's skill in that area would not be of such importance on a well-coached team featuring a versatile offense.

Next, it is painfully obvious that Salmons is a better all-around player than Gordon, and that his other skills (especially defense) are extremely important.

I could go on, but really, given the deeply flawed premise, I hardly see the point. It is sadly ironic, though, that the author's "desire to find out the truth" led him to use such a poor premise from which to mount his argument.
#14891
Re:Too much one-on-one
May 16 2009 13:54:32
badnewsintennisshoes wrote:
This is a classic example of the dangers of over-focusing on a narrow set of statistics. In every sport, countless people seek the magic number – or numbers – to be used as a shortcut. But the truth is that no narrow set of numbers can ever substitute for the more sophisticated analyses necessary to make good personnel decisions. And, in many cases, such as this one, the numbers are actually misleading.

What method of analysis would you choose to see if Gordon goes one on one too much?

With regards to this particular statistical approach, it is obviously flawed, and is a ridiculous way to compare Gordon and Salmons (or any other players). According to the above list, Salmons, Randy Foy and Nate Robinson are all more "effective" one-on-one players than Carmelo Anthony. Do I really have to elaborate further?

It shows that they were more effective in their attempts, I think we all realize that the more attempts someone has and the type of defender that they are going up against would sway such a number.

However, Gordon clearly draws among the best defenders on the other team and does not have a small sample size of attempts, so his effectiveness would seem quite legitimate.

Furthermore, I'd sure like to know how "one-on-one" is defined. Is there necessarily dribbling involved, or is a contested jump shot considered as well? In other words, let's see some definitions so that we can take into account how the posted (and clearly flawed) numbers were generated.

I believe dribbling is always involved. There is a "catch and shoot" category which would define the times a player is passed the ball and shoots right away. This is the video scouting service used by every NBA team by the way.

The way that players perform is also obviously dependent in large part on how they are coached. In the context of the Bulls, VDN was so atrocious that to judge players on their raw numbers (even accurate and meaningful ones) while they played in such a poorly designed offensive is silly. In other words, the biggest single step the Bulls could take would be to replace VDN, irrespective of the players in uniform. A better coach, for example, would not resort to predictably having Gordon go one-on-one (or two) at the end of tight games. The point being that Gordon's skill in that area would not be of such importance on a well-coached team featuring a versatile offense.

This would improve my case that Gordon does not go one on one too much.

Next, it is painfully obvious that Salmons is a better all-around player than Gordon, and that his other skills (especially defense) are extremely important.

Actually, not only do I not think this is painfully obvious, but I think it's painfully obvious the reverse is true.

I could go on, but really, given the deeply flawed premise, I hardly see the point. It is sadly ironic, though, that the author's "desire to find out the truth" led him to use such a poor premise from which to mount his argument.

You haven't really described how the premise is flawed in any way that goes against the point I'm making. I agree there are considerations to look at when checking out these facts, but most of those considerations back up my argument further rather than lessen it (like your use of VDN and the Bulls offense).
#14892
Too much one-on-one
May 16 2009 14:34:00
I agree that Gordon's statistics would improve if he was not relied upon so much to create his own shot (i.e. if the Bulls had a better coach and more diverse offense). My main points are these:

Defining Gordon’s quality as a one-on-one player is not the right way to go about deciding whether or not he should be retained.

Using this set of statistics is clearly not a good way to define one-on-one quality in any case, which is why the premise is flawed. Presumably you have watched plenty of basketball, correct? Well, then can you please explain to me how this statistic could possibly be an accurate expression of how "effective" a one-on-one player is when, in that context, Salmons, Randy Foy and Nate Robinson are all more "effective" one-on-one players than Carmelo Anthony? I mean your whole argument about Gordon vs. Salmons is based on this set of statistics, so by the very same measure, Salmons is somehow more valuable than Carmelo. Do you actually believe that to be the case?

There's no doubt that Gordon goes one-on-one too much, but that is, as always, mainly a coaching problem. If the Bulls had a good coach, and Gordon was made to limit his one-on-one play, the team would benefit. But that, of course, is only one small part of the equation needed to assess whether he should be kept on the team, or what his overall value is to the team.

With regards to Salmons vs. Gordon, the former is a better rebounder, better defender, averages more steals and less turnovers, and there is “statistically” little between them offensively. How, exactly, do you arrive at the conclusion that Gordon is a better all-around player?
#14893
Re:Too much one-on-one
May 16 2009 16:07:46
badnewsintennisshoes wrote:
With regards to Salmons vs. Gordon, the former is a better rebounder, better defender, averages more steals and less turnovers, and there is “statistically” little between them offensively. How, exactly, do you arrive at the conclusion that Gordon is a better all-around player?

Salmons has done it for one year. Gordon has done it for 5, 3 at a stellar rate.

Salmons turnover rate has been significantly higher than Gordon's every year but this one where it was 11.7% vs. 11.9%.

Their steal rates are almost negligible.

Rebounding rates are in large part a difference in position. Salmons played most of the year at SF, a position which produces a lot more rebounding attempts than SG. Also, Salmons was a poor rebounder for a SF at that.

Finally, there is statistically a WHOLE LOT of difference between them. Salmons has 1 year, ONE where he performed like a starting caliber player. And even in that 1 year Ben still average 3 more pts per 36mins than Salmons... and yes, that is a significant difference.
#14897
Re:Too much one-on-one
May 16 2009 16:27:21
Love the article Doug. However just got a few quick questions. Do these stats exclude the times that double teams came thus no longer one on one? And if they don't, do you have stats that can show in these one on one situations if the double team came what the player did whether still forced a shot or passed the ball and whether this pass lead to a score for the team? How about what percentage of the time going one on one lead to a turnover?
#14900
Too much one-on-one
May 16 2009 16:27:50
"Salmons has done it for one year. Gordon has done it for 5, 3 at a stellar rate."

I agree that there is a difference, but to dismiss Salmon's best year as an anomaly is a highly questionable strategy.

"Rebounding rates are in large part a difference in position."

Do you seriously doubt that Salmons is a much better rebounder, irrespective of position played? C'mon.

"Ben still average 3 more pts per 36mins than Salmons... and yes, that is a significant difference."

Please, three points is significant? In the likely event that Salmons were to make two or three defensive stops per game that Gordon wouldn't, that "significant" difference would be reversed. Three points is trivial if one of the players is a better - and more versatile – defensive player. That is the case with Salmons.
#14901
Too much one-on-one
May 16 2009 16:49:58
This conversation has also taken on a bit of a misleading direction. No one – including the original poster, I believe – is focusing mainly on some zero sum question of whether the Bulls should keep Salmons or Gordon. Although given all of the factors, including salaries and overall value to the team, I would choose Salmons, the more pertinent question is just how much does Gordon contribute to the team?

As some of my previous points suggest, Gordon has only one true strength – scoring – and even that is not as overwhelmingly impressive as many seem to think.
#14903
Re:Too much one-on-one
May 16 2009 17:12:56
badnewsintennisshoes wrote:
Defining Gordon’s quality as a one-on-one player is not the right way to go about deciding whether or not he should be retained.

Not remotely the purpose of the article nor a point made in the article. The point was made if you feel we should remove Gordon due to him going one-on-one too much that Salmons isn't the solution to that problem as he stops the ball more frequently with less positive results. If you want to remove Gordon for different reasons then this article is not intended to address those.

Using this set of statistics is clearly not a good way to define one-on-one quality in any case, which is why the premise is flawed. Presumably you have watched plenty of basketball, correct? Well, then can you please explain to me how this statistic could possibly be an accurate expression of how "effective" a one-on-one player is when, in that context, Salmons, Randy Foy and Nate Robinson are all more "effective" one-on-one players than Carmelo Anthony? I mean your whole argument about Gordon vs. Salmons is based on this set of statistics, so by the very same measure, Salmons is somehow more valuable than Carmelo. Do you actually believe that to be the case?

There is a difference when you add an extreme amount of attempts at going one-on-one in the difficulty of those attempts and there is a difference in the coverage assigned to different players. Gordon clearly gets the best defenders drawn on him and double teams ready to take him every night, so this doesn't seem to be a negative in his stats though whereas Foye/Robinson may have better stats than warranted if they were focal points.

There's no doubt that Gordon goes one-on-one too much, but that is, as always, mainly a coaching problem.

I agree that it's up to the coaches, but my data shows that Gordon goes one-on-one no more than any scorer of his ability level.

With regards to Salmons vs. Gordon, the former is a better rebounder, better defender, averages more steals and less turnovers, and there is “statistically” little between them offensively. How, exactly, do you arrive at the conclusion that Gordon is a better all-around player?

Salmons isn't a better rebounder. He averages .7 more rebounds than Gordon per minute this season while playing a much higher rebounding position. To put it in perspective Luol Deng at the same position in the same offense averaged 2.3 more rebounds per 36 minutes than Salmons. So no, I don't believe Salmons is a "better" rebounder. If he is, it's by a very slim margin.

Moreover, the gap in all of the things you mention are barely noticeable at all. I don't know that I would say Gordon is a better all-around-player in that he has a wider variety of skills he executes at a high level. However, I think he's a better player due to overall package he brings to the team vs what Salmons brings. His skills are more focused in scoring, but scoring is a huge array of skills that people lump into one category.
#14904
Re:Too much one-on-one
May 16 2009 17:15:41
wjb1492 wrote:

See, but now you're changing the question. I looked at points only and compared him to that group because a question had come up in the context of scoring - Ben is an elite scorer, so that's best group to compare him to when discussing variations in points scored. Obviously, when you start adding other stats into the equation, Ben is going to rate differently than those other players. I would absolutely agree that he's a bit farther down the list as an all-around player - that is, he scores roughly "as much" as other 20+ ppg guy, but many of them contribute more in other categories. And comparing Ben to max contract type guys for all-around game is no longer the correct comparison group - if Ben had the all-around talent of these other guys we wouldn't be arguing about whether he's worth $8M or $10M per year.

If you want to get to the issue of variation, you can't just look at how many games fall below a certain constant threshhold. Ben's average gamescore and st.dev. for the season were 13.59 and 7.68 (I used gamescores in a teamwide analysis of player "great games" and "awful games" and how those related to team record in another post over on BaB, so have those for Bulls players but not anyone else in the NBA), so it is absolutely to be expected that he'd have more games below 10 than another player who's average gamescore was higher to begin with - but at this point you haven't addressed consistency at all - is that 7.68 standard deviation excessively high, or is it right in line with other players of Ben's quality?


Ben is an ELITE scorer? So there are nearly 27-30 elite scorers in the league? I would think you'd use the word ELITE with the top echelon of scores not about 20% of the starters in the league. He's a GOOD scorer, if he didn't have nearly as many bad nights as he did he'd be close to be an elite scorer in this league... which he's not.

He's just happened to be the best scorer on the Bulls for a few years... which will be overtaken by Rose in the next couple of years.
#14905
Re:Too much one-on-one
May 16 2009 17:16:30
badnewsintennisshoes wrote:
This conversation has also taken on a bit of a misleading direction. No one – including the original poster, I believe – is focusing mainly on some zero sum question of whether the Bulls should keep Salmons or Gordon. Although given all of the factors, including salaries and overall value to the team, I would choose Salmons, the more pertinent question is just how much does Gordon contribute to the team?

As some of my previous points suggest, Gordon has only one true strength – scoring – and even that is not as overwhelmingly impressive as many seem to think.


First, you should check out the forums rather than posting from the comment form, I think it's easier (not that you have to, but you might like it more).

I do agree with you that this isn't intended to be a Salmons vs Gordon article. I just related Gordon to Salmons in one on one usage since many people complain about Gordon's one on one usage and want Salmons.

If I had to choose between Gordon and Salmons I'd take Gordon, if his contract got too high then I'd let him walk. I just think Salmons is a short term option. He's a solid player, but I don't think there are more then 2-3 better shooters in the NBA than Gordon (not overall scorers, but pure shooters), and having him next to Rose is something I value a lot due to the spacing he brings.

I'd like to see the coaching staff alter their use of Gordon to run him off a lot more screens, and I'd like to see Gordon improve his shooting coming off screens. However, I think we're unlikely to have another shooter as good as Gordon again if we let him go, and having him space the floor for Rose is really nice.

If he demands too much money I'd pack his bags too, Gordon certainly has flaws in his game.
#14906
Too much one-on-one
May 16 2009 19:48:53
I'd say, Doug, that we are close to refining our differences.

We agree that Gordon plays too much one-on-one, and that better coaching could resolve that. We differ on how good, and how important an offensive player Gordon is. We also apparently disagree about his other skills, as I consider him to be average to below average as a rebounder, passer and defensive player.

Given his likely contract requirements, I think he should be moved on. There are plenty of excellent shooters who could fill the void, especially if the changes are made to the offense which we agree would be helpful.

I never meant to suggest that Salmons should be kept at all costs, but he is a very solid and versatile player. And, given the big difference in their contracts, I would much prefer to see him retained.

Finally, the reason that I jumped in rather aggressively, is that I become impatient when statistics - especially questionable ones – are used as a foundation for an argument. You've expounded on your original post, and I now see that you aren't really putting much emphasis on the stats that you used initially.
#14916
Re:Too much one-on-one
May 16 2009 21:15:12

Ben is an ELITE scorer? So there are nearly 27-30 elite scorers in the league? I would think you'd use the word ELITE with the top echelon of scores not about 20% of the starters in the league. He's a GOOD scorer, if he didn't have nearly as many bad nights as he did he'd be close to be an elite scorer in this league... which he's not.

He's just happened to be the best scorer on the Bulls for a few years... which will be overtaken by Rose in the next couple of years.


I think he's better than good. I'd say he's a great scorer. I agree that I wouldn't go with 'elite' scorer though. As for Rose overtaking him, I think it's highly unlikely Rose will ever score with Gordon's combination of volume and efficiency.
#14926
Re:Too much one-on-one
May 16 2009 21:21:14
badnewsintennisshoes wrote:
We agree that Gordon plays too much one-on-one, and that better coaching could resolve that. We differ on how good, and how important an offensive player Gordon is. We also apparently disagree about his other skills, as I consider him to be average to below average as a rebounder, passer and defensive player.

I think he's average in those areas except rebounding where he's probably a bit below average, I feel Salmons is also deficient in that area though which is why I said he's no better.

Given his likely contract requirements, I think he should be moved on. There are plenty of excellent shooters who could fill the void, especially if the changes are made to the offense which we agree would be helpful.

There really aren't plenty of excellent shooters, because of Gordon's combination of quick release and ability to hit a 1 dribble jumper if you buy a fake it's really hard to find someone who demands his attention, because if you close out on him wildly he can easily get past you and score on you where most other good shooters really can't do that.

As for his contract requirements, I think instead of worrying about what his requirements are you make him an offer that you think is good for your team and see how it goes. If it doesn't work out that's fine, but this is going to be nuclear winter for FAs. The economy is going to sink the available money, and their aren't tons of teams with cap space anyway. Gordon isn't going to go at nearly teh value he would in a different year. That's largely the reason I want to sign him so badly.

I think you can get Gordon for under 9 million a year this summer. At that pricetag he's a bargain as there are very few plays who aren't on rookie contracts who make less than that and are better than Gordon.

I never meant to suggest that Salmons should be kept at all costs, but he is a very solid and versatile player. And, given the big difference in their contracts, I would much prefer to see him retained.

I actually don't think there will be that big a difference in their contracts for the next 2 years. I think if Gordon gets 8 million a year you're paying him 4th option money. He's at least a 4th option and maybe a 3rd option on a great team, so at that pricetag I think you've set yourself up for something good with his deal.

Salmons is a short term solution who doesn't have a great track record. I don't think he'll fall apart, but keeping him isn't building towards anything, but holding down the fort for a year or two. Getting Gordon is definitely building towards something, but you can argue it's building towards something bad I suppose. The problem is that we're going to be hard pressed to get a better player than Gordon for the same salary space at a later point.

Finally, the reason that I jumped in rather aggressively, is that I become impatient when statistics - especially questionable ones – are used as a foundation for an argument. You've expounded on your original post, and I now see that you aren't really putting much emphasis on the stats that you used initially.

The main use of the stats was just to show that Gordon does not go to the isolation with any great frequency relative to other good scorers around the league. That was the only point I was really aiming to make.
#14928
Re:Too much one-on-one
May 16 2009 22:16:21
i don't know if these are really "questionable" stats, they're very true, it's just like one of those extra baseball stats
#14938
Re:Too much one-on-one
May 17 2009 02:28:41
dougthonus wrote:

Ben is an ELITE scorer? So there are nearly 27-30 elite scorers in the league? I would think you'd use the word ELITE with the top echelon of scores not about 20% of the starters in the league. He's a GOOD scorer, if he didn't have nearly as many bad nights as he did he'd be close to be an elite scorer in this league... which he's not.

He's just happened to be the best scorer on the Bulls for a few years... which will be overtaken by Rose in the next couple of years.


I think he's better than good. I'd say he's a great scorer. I agree that I wouldn't go with 'elite' scorer though. As for Rose overtaking him, I think it's highly unlikely Rose will ever score with Gordon's combination of volume and efficiency.


Fair enough if you want to quibble with that choice of words. When you consider points scored and efficiency, I'm comfortable going with elite. If that's going to distract people from the main point, we can go with very good instead.

However, the point is that if you want to discuss whether Ben is actually inconsistent in points scored from night to night, then other SGs that score 20+ ppg is the appropriate comparison group on that issue. And Ben is roughly equal to those other 20+ ppg SGs in his variation in points from night to night. He's simply not having more bad scoring nights than other guys right at his scoring level.

You can easily conclude he's not Kobe or D-Wade - but no one's putting him up for MVPs or a max contract. You can expand the issue to other areas of performance - but that's why he's looking at a much lower contract than those other 20+ ppg SGs have or are expected to get.

What I don't think you should keep doing is running around shouting about Ben being inconsistent, implying that he's more inconsistent than other high scoring SGs - because the numbers show he isn't. No matter how you fixate on his few low scoring games, it's not going to change the fact that the other guys have a similar number of bad nights.

Kobe's and D-Wade's aren't [i]as[/] bad on a straight points basis, but that's because their averages are higher to start with - in other words, a 21 point night for D-Wade is roughly the same as a 180 point night from Kobe is roughly the same as a 13 point night from Ben in terms of variation from their averages, all being right around 1 standard deviation below their average.
#14951
Re:Too much one-on-one
May 17 2009 02:30:10
dougthonus wrote:

I actually don't think there will be that big a difference in their contracts for the next 2 years. I think if Gordon gets 8 million a year you're paying him 4th option money. He's at least a 4th option and maybe a 3rd option on a great team, so at that pricetag I think you've set yourself up for something good with his deal.


I think saying Ben Gordon is maybe a third option is selling him short. He's easily a third option and put up better scoring numbers than nearly every playoff team's second option in terms of volume and efficiency. Of the teams left in the running, he put up better numbers than Gasol, Billups, Artest, Allen, Lewis or Williams. Obviously some of those guys are better at other aspects of the game than Gordon is, but as a second scoring option on a championship team, I think Gordon is certainly good enough.
#14952
Re:Too much one-on-one
May 17 2009 02:43:21

I think saying Ben Gordon is maybe a third option is selling him short. He's easily a third option and put up better scoring numbers than nearly every playoff team's second option in terms of volume and efficiency. Of the teams left in the running, he put up better numbers than Gasol, Billups, Artest, Allen, Lewis or Williams. Obviously some of those guys are better at other aspects of the game than Gordon is, but as a second scoring option on a championship team, I think Gordon is certainly good enough.


Third option is a bad phrase, third best player would be better.

If you look at recent title teams the Spurs have 3 guys better than Gordon. You could argue the Pistons had 3 guys better than Gordon when they won. The Bulls may or may not have had 3 guys better than Gordon depending on what you feel about Rodman/Grant/Kukoc etc...

Of the teams that win titles, Gordon would be the 3rd best player on about half and the fourth best player on half I'd say.
#14953
Re:Too much one-on-one
May 17 2009 02:56:29
badnewsintennisshoes wrote:
I agree that Gordon's statistics would improve if he was not relied upon so much to create his own shot (i.e. if the Bulls had a better coach and more diverse offense). My main points are these:

Defining Gordon’s quality as a one-on-one player is not the right way to go about deciding whether or not he should be retained.

Using this set of statistics is clearly not a good way to define one-on-one quality in any case, which is why the premise is flawed. Presumably you have watched plenty of basketball, correct? Well, then can you please explain to me how this statistic could possibly be an accurate expression of how "effective" a one-on-one player is when, in that context, Salmons, Randy Foy and Nate Robinson are all more "effective" one-on-one players than Carmelo Anthony? I mean your whole argument about Gordon vs. Salmons is based on this set of statistics, so by the very same measure, Salmons is somehow more valuable than Carmelo. Do you actually believe that to be the case?

There's no doubt that Gordon goes one-on-one too much, but that is, as always, mainly a coaching problem. If the Bulls had a good coach, and Gordon was made to limit his one-on-one play, the team would benefit. But that, of course, is only one small part of the equation needed to assess whether he should be kept on the team, or what his overall value is to the team.

With regards to Salmons vs. Gordon, the former is a better rebounder, better defender, averages more steals and less turnovers, and there is “statistically” little between them offensively. How, exactly, do you arrive at the conclusion that Gordon is a better all-around player?


I agree with badnews on the stat thing being flawed only because we don't have definitions. I don't agree about Gordon being moved, he is great for Rose. Stats in Basketball are not as advanced as Baseball and we have to be careful so we can't base everything to be true based on the stats, we have to use them to back up our own observations. There is now way I'd rather have Nate Robinson taking the last shot over Carmelo and I already saw BG and Lebron are equal, no way. I already listed making it to the FT line also if they make 1 vs 2, plus what is one on one, does that count double teams?, what about triple teams which Gordon likes to do a lot. What if they are in a zone and he tries to shoot in between 2 guys running at him?

In this case however, I think that Doug is basing these statistics on his own observations and beliefs therefore they are one aspect to bolster his arguement. During the thread he has posted other stats to back up his point as well so I think the article was supposed to be about one set of stats so I don't have a problem with it.
#14955
Re:Too much one-on-one
May 17 2009 03:06:58
As far as Salmons vs Gordon going one on one I think I would rather have Gordon slightly but people are sleeping on JFish. I watched a lot of the Sacramento games because I am up late on the East Coast a lot and watch the West Coast games on League Pass. He was a consistant scorer on that team all year with a changing lineup and a terrible team, then got traded to a better team with more scoring and stepped right in. He also comformed right away to the crappy offense and learned how to score in a spot where Deng was complaining. Deng is obviously terrible going one on one. Therefore I think we have to give Salmons more credit, no I don't think he is more important than Gordon but he might be slightly better overall than Gordon. I know Ben has done it for more years so I will give him the benefit of the doubt but Salmons is not far off.

Also Salmons being a career -7 means very little, he came to the Bulls and was the biggest difference in the trade. Miller helped as well but Salmons was a 20 point scorer before the injury and the Bulls went on to have a great record after the trade so his +/- couldn't have been that bad right or was it? It is only good for the extremes, in this case who cares.
#14958
There are too many comments to list them all here. See the forum for the full discussion.
You need to login or register to post comments.
Discuss this item on the forums. (83 posts)